Petreley.com

Because Nobody Does It Better

It’s time we started paying attention to what Nobody is doing. Nobody rarely gets the attention he deserves yet, strictly speaking, Nobody is not an unsung hero of our times. You may recall the song by Carly Simon hailing Nobody as the best of the best.

Nobody does it better
Makes me feel sad for the rest

Nobody should be a prominent figure in Linux and open source. Even Microsoft’s own Steve Ballmer has such respect for Nobody that he invited him by name to attend his keynote at the Gartner Symposium and ITxpo, 2004. “Nobody should come to this keynote,” said Ballmer. (transcript)

What else does Steve Ballmer have to say about Nobody? He acknowledge’s Nobody’s wisdom in selection of software, “Nobody pays for software on Linux.” (News.com) And Steve Ballmer regards this Nobody as Microsoft’s greatest competitor, admitting that “Nobody out there has the talent pool that we have.” (Business Week)

While Ballmer may appear to threaten those who adopt Linux with patent lawsuits, at one point he let it slip that, “When it comes time to get indemnification for intellectual property on Linux, Nobody will give it to you.” Furthermore, Steve Ballmer credits this same hero as the safety net for those who may be concerned about future security issues in Linux, “There’s Nobody to hold accountable for security issues with Linux.” (Entmag)

What about Windows security? This is what Ballmer has to say about the person Nobody for whom he has so much respect, “Nobody doubts today that it was a good idea to make a TCP/IP stack part of Windows,” ostensibly because putting TCP/IP into Windows connects the highly vulnerable system to the Internet. (Entmag)

So there you have it. What more can we add to Steve Ballmer’s open praise for Nobody and his admissions of Linux superiority and Windows inferiority? Nobody could make a better argument for the adoption of Linux and abandonment of Windows. Unfortunately, he was not available for comment.

Microsoft is still attempting to use any means necessary, including behind-closed-doors politics, to get ISO to crown OOXML as an official standard. The only thing more mysterious than the methods Microsoft is attempting to use to accomplish its standardization goals is why anyone cares?

Think about this for a moment. What is the most practical reason to establish a standard? The primary reason a standard is important is to enable seamless access to the standards from disparate sources. In most cases, we’re talking about standard data formats for applications from disparate sources. Web standards encourage people to create web sites you can navigate using Microsoft Internet Explorer, Mozilla Firefox, Opera, Konqueror, Safari, or some other browser and get basically the same experience.

Now assume for a moment that OOXML became an official ISO standard. What disparate applications would access the OOXML data? OOXML is nothing more than a 6,000 page description of “how Microsoft Office works and how it handles data”. The only way to observe this standard fully would be to create a perfect clone of Microsoft Office, including even the bugs in various Microsoft Office versions.

No such clone exists, nor is any such clone ever likely to exist. Therefore, even if OOXML is declared an ISO standard, it fails to fulfill the most practical purpose of a standard because we do not have a reasonable choice of applications which observe the standard.

Why is it that no such clone is ever likely to exist? Because the so-called standard is not written with the purpose of encouraging third parties to create a Microsoft Office clone. The so-called standard is nothing more than a description of “how we did it”.

One might argue that Linux Standard Base (LSB) follows a similar philosophy. LSB is pretty much a description of “how a basic subset of Linux works”. Like OOXML, LSB is very detailed and complex. But here’s the important difference. Several Linux distributions support LSB, which gives it the aforementioned practical value of being supported by disparate sources.

LSB would be like OOXML only if LSB described “exactly how Ubuntu 7.10 works, including even the bugs in Ubuntu 7.10″. If that were the case, then the only hope of supporting LSB would be to create a perfect clone of Ubuntu 7.10. In this case, LSB would have no value as a standard.

Microsoft is not the only problem

It is fairly obvious why Microsoft wants the ISO to approve OOXML as a standard. Government agencies, among other customers and potential customers, are getting more savvy about openness. Some are adopting policies that preclude the use of closed, non-standard data formats. Microsoft must find a way to keep these people locked into using Microsoft Office, so it needs to get its document format declared an open standard.

Now assume once again that OOXML is declared an ISO standard. Anyone who changes his or her position on OOXML because of its status with the ISO isn’t thinking clearly and is contributing to the absurdity of OOXML as a standard. It is not enough to be able to put a check in the “open standards” box in order to approve of adopting OOXML in their organization.

Anyone who thinks, “Hey, we can adopt OOXML now because ISO declares it is an open standard,” has lost touch with the practical reason for adopting open standards to begin with. In other words, anyone with their head firmly attached shouldn’t care what the ISO says about OOXML. OOXML has no real value as a standard until a third party application adequately supports OOXML.

But wait, there’s more

It is equally important that the third party application supports OOXML without having to license anything from Microsoft. If any third party depends on Microsoft for its support of this so-called standard, the standard no longer has any practical value. As long as Microsoft builds into its so-called standard the power to eliminate competitors, the practical value of having disparate sources supporting said standard is nil. There is no practical value of having disparate sources support a standard if, with the flick of a patent claim or lawsuit, Microsoft can eliminate those other disparate sources.

In conclusion, anyone who really understands the value of openness and open standards doesn’t even need to pay attention to the soap opera saga of OOXML and the ISO. For the foreseeable future, OOXML should be off limits based on principle alone, no matter what the ISO decides.

Trend Micro is trying to enforce a patent in order to collect money from competitors. One of the latest targets is Barracuda Networks. From the Barracuda Networks press release:

Campbell, Calif., Jan. 29, 2008 – Barracuda Networks Inc., the worldwide leader in email and Web security appliances, today announced it plans to defend itself, the open source community and the free and open source Clam AntiVirus software from a patent threat by Trend Micro. Barracuda Networks’ decision to take action comes after repeated requests from Trend Micro demanding Barracuda Networks remove ClamAV from its products or pay a patent license fee.

“Trend Micro’s actions illustrate that ClamAV and other open source projects remain vulnerable to commercial patent holders attempting to unjustly hinder the free and open source community,” said Dean Drako, president and CEO of Barracuda Networks. “Trend Micro appears to be seeking an interpretation of its ‘600 patent such that it would have exclusive control of gateway antivirus scanning. Scanning for viruses at the gateway is an obvious and common technique that is utilized by most businesses worldwide. So this interpretation would mean that anyone, including the owners of the more than one million active ClamAV installations, could potentially be sued by Trend Micro.”

The basic idea is that Trend Micro as a patent on detecting and filtering viruses at the Internet/e-mail gateway. ClamAV filters viruses at the gateway, therefore ClamAV allegedly violates the patent.

Barracuda Networks believes the patent is invalid due to prior art. That’s probably true, but it is irrelevant to the point I’d like to make. This patent, like so many others, is idiotic and should never have been granted.

The “Well, duh!” Test

This patent is typical of a group of patents that should all be filed under “absurd”. The easiest way to identify such patents is with the “well, duh!” test. Ask the fundamental question about the basis for the patent, such as “Does it make sense to scan emails for viruses as the emails arrive at the gateway?” The answer is clearly, “well, duh!” Whenever you get this answer, the patent should not be awarded.

Let’s expand on this just a bit to show when a patent application would fail the “well, duh!” test.

1. Q. Would it be cool to place speakers around the room to get a surround sound effect?

A. “Well, duh!” Patent denied.

2. Q. Would it be cool to be able to emulate surround sound with only two main speakers?

A. “Well, duh!” Patent denied.

3. Q. Is it cool to emulate surround sound with two speakers by starting with real-time analysis of the sound using a real-time fast-fourier transform on each of the 5 or 7 channels… etc?

A. “A fast what?” Patent remains under consideration.

The difference should be obvious. The third example takes REAL INTELLIGENCE and REAL WORK. Any idiot can figure out that it’s a good idea to create surround-sound with front speakers. It takes work and intelligence to come up with an idea on HOW to create a surround sound effect using only front speakers.

If anyone actually gets a patent on a common-sense good idea, it will be because they happened to get it to the patent office before anyone else could, and before anyone could establish prior art based on the same common-sense idea.

In short, the patent on HOW to create surround sound with two speakers is awarded based on a brilliant idea. It took intelligence, probably required some education and almost definitely required research. If a patent is awarded, it rewards this intelligence, research, etc. If a patent is awarded on the common-sense ideas, then what is rewarded is speed. The person who got the patent simply won a race to the patent office. Does this person or the company he represents actually deserve to receive a share of the profits made on products that implement these common-sense ideas? I argue that he/the company does not.

The Barracuda Case Argues for Patent Reform

In this author’s opinion, the Barracuda case is not about open source, even if it is an attack on open source. This case is yet another case that argues for patent reform. It is time for an overhaul of patent practices, especially software-related patents.